The goal of this proposal is to define HIV-1 genes that determine the infectiousness of the virus. Our studies in injection drug users indicate that the range of HIV-1 variants that are readily transmitted demonstrate restricted genomic variation when compared to that observed in prevalent seropositive subjects. Available data, including our own in injection drug users, indicate that transmitted viruses are derived from a minority population of the pool of non-syncytia inducing HIV-1 variants present in the donor. These data suggest that determinants other than Sl/NSI phenotype influence transmissibility of virus. Studies in our laboratory with recombinant molecular clones also indicate that mutations within the V3 region that affect viral tropism are not sufficient to account for the marked differences in infectiousness observed among different viral variants. Studies of inoculation of mice with severe combined immunodeficiency transplanted with human peripheral blood mononuclear cells (Hu-PBL-SCID mice) have demonstrated a pattern of transmission similar to that observed in recently infected humans; infection is preferentially established with NSI variants of restricted clonality. This finding is in marked distinction to the pattern of infection in cultured PBMC, in which infection with Sl variants predominates. In the current proposal, chimeric clones will be created between the molecular clone, NLHX-ADA-gg, and the biological clone, HIV MN, which differ markedly in their ability to establish infection in Hu-PBL-SCID mice. The clones will initially include large fragments (approximately 2000 bp) from the poorly infectious HIV-1 MN inserted into NLHX-ADA-gg, using unique restriction sites that have been identified in NLHX-ADA -gg. The infectiousness of the chimeric clones will then be evaluated in Hu PBL-SCID mice. When a region that reduces infectiousness of NLHX-ADA-gg is identified, subregions of that region will be inserted into NLHX-ADA-gg to more specifically identify the region influencing infectiousness. In all cases regions will be identified that influence infectiousness in Hu-PBL-SCID mice, but do not affect the ability of the virus to establish infection in cultured PBMC. This will ensure that the observed effects are not due to an impact on the fundamental viability of the virus. Once a region(s) influencing infectiousness is identified using the Hu-PBL-SCID mice, that region will be sequenced from amplified products from two groups of injection drug users. The first is a group of recently infected (p24 positive, seronegative) subjects identified as part of an ongoing study of acute infection. The other is a group of prevalent seropositive injection drug users, also identified as part of an ongoing study, who show at least 10% variation in the env region. Sequences from these two groups will be compared for motifs that are shared or are divergent from motifs found in the regions responsible for infectiousness in the Hu-PBL-SCID mice. These studies provide an opportunity to identify gene products for infectiousness in the Hu-PBL-SCID mice. These studies provide an opportunity to identify gene products that optimize infectiousness, thereby providing potential antigens for use in candidate anti-HIV vaccines. These studies should also afford important insights into the pathogenesis of HIV-1 infection.